Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A communication system, comprising: a plurality of microphones; a device detection and localization system comprising, within a same physical housing, a radio frequency identification tag reader; a communication server, including: memory; a processor; a conference application stored on the memory and executed by the processor, wherein through execution of the conference application the communication server: determines a location of a first party providing first audible information to the plurality of microphones, wherein the location of the first party is determined based on different arrival times of the first audible information at two or more of the plurality of microphones; identifies a first identification device at the location of the first party providing the first audible information, wherein the first identification device is a personal communication device of the first party that comprises, in the same physical housing, a radio frequency identification tag, and wherein the first identification device is identified based on a caller ID of the first party and the radio frequency identification tag in the first identification device; correlates the caller ID of the first identification device to the first party; provides the first audible information as part of a conference call; and generates an output signal identifying the first party providing audible information in the conference call.
A communication system identifies speakers in a conference call. It uses multiple microphones to capture audio. A device detection system, housed with an RFID reader, locates participants. A communication server, running a conference application, determines a party's location by analyzing the arrival times of their voice at the microphones. It identifies a personal communication device (like a phone with a built-in RFID tag) at that location by matching the caller ID and RFID tag. The system then links the caller ID to the person's identity and outputs a signal that identifies the speaker during the conference call.
2. The system of claim 1 , further comprising: a feature server, wherein the first audible information and the output signal identifying the first party are provided to the feature server.
The system described in the previous claim about identifying speakers in a conference call also sends the audio and speaker identification information to a feature server. This feature server can then use this information for additional functionalities, such as generating meeting transcripts, performing sentiment analysis on the speaker's voice, or providing real-time speaker labels during the conference.
3. A method for identifying a party to a communication session, comprising: receiving, by a processor, a first audible signal at a first communication device; determining, by the processor, a location of a source of the first audible signal; determining, by the processor, a location of a first identification device, wherein the location of the first identification device is determined based on different arrival times of a signal from the first identification device received at a plurality of sensors; determining, by the processor, that the location of the source of the first audible signal corresponds to the location of the first identification device, wherein the first identification device is a personal communication device of the first party that comprises, within the same physical housing, a radio frequency identification tag, and wherein the first identification device is identified based on a caller ID of the first party and the radio frequency identification tag in the first identification device; determining, by the processor, an identity of a first party associated with the first identification device based on the caller ID of the first party; providing, by the processor, the first audible signal as part of a conference call; and generating, by the processor, an output identifying the first party as the party from whom the first audible signal is received in the conference call.
A method identifies speakers in a conference call. A processor receives audio. It determines the location of the audio source. It also determines the location of a personal communication device (like a phone with a built-in RFID tag) by analyzing the arrival times of its signal at multiple sensors, identifying the device via caller ID and RFID tag. The method confirms that the audio source and the communication device are at the same location. It identifies the person associated with the device, using the device's caller ID and provides the audio as part of a conference call, generating an output that identifies the speaker.
4. The method of claim 3 , wherein determining an identity of the first party comprises: receiving stored information identifying the first party from the first identification device.
The method for identifying a speaker in a conference call, which involves determining the location of an audio source and matching it to a personal communication device's location as described in the previous claim, identifies the speaker by receiving stored information directly from the communication device.
5. The method of claim 4 , wherein the stored information is read from the radio frequency identification tag and the location of the radio frequency identification tag is determined by a radio frequency identification reader.
In the method where speaker identification uses stored information from the communication device, as detailed in the previous claim, that stored information is read from an RFID tag. An RFID reader determines the location of that RFID tag.
6. The method of claim 4 , wherein the first identification device comprises a second communication device, wherein the second communication device comprises a first near field communication system, and wherein the stored information is read from the second communication device and the location of the second communication device is determined using signals passed between the first near field communications system and a second near field communication system.
In the method where speaker identification uses stored information from the communication device, as described two claims ago, the communication device contains a near-field communication (NFC) system. The stored information is read using this NFC system, and the location of the communication device is determined using signals exchanged between this NFC system and another NFC system.
7. The method of claim 4 , wherein the stored information identifying the first party received from the first identification device associated with the first party is applied to access personnel identification data, wherein at least some of the personnel identification data is entered during a registration step and is stored as user identification data, and wherein a name of the first party is obtained from the personnel identification data.
In the method where speaker identification uses stored information from the communication device, as outlined three claims ago, the information received from the communication device is used to access personnel identification data. This data, including the person's name, was entered during a registration process and is stored as user identification data.
8. The method of claim 4 , further comprising: receiving a second audible signal at the first communication device; determining a location of a source of the second audible signal; determining a location of a second identification device; determining that the location of the source of the second audible signal corresponds to the location of the second identification device; determining an identity of a second party associated with the second identification device; and generating an output identifying the second party as the party from whom the second audible signal is received.
Building on the method for identifying a speaker, as described four claims ago, the method also identifies a second speaker. It receives a second audio signal, determines its source location, and determines the location of a second communication device. If the audio source and device locations match, it identifies the person associated with that second communication device and outputs a signal identifying the second speaker.
9. The method of claim 8 , wherein the first and second audible signals are received simultaneously.
In the method that identifies both a first and second speaker, as detailed in the previous claim, the audio signals from both speakers are received at the same time. This allows for simultaneous identification of multiple speakers in a conversation.
10. The method of claim 4 , further comprising: displaying the output identifying the first party as the party from whom the first audible signal is received to a party using a second communication device, wherein the first communication device and the second communication device are participating in an active communication session.
Expanding on the method of identifying a speaker using location and device information, described five claims ago, the output identifying the speaker is displayed to another participant in the conference call using their communication device. This provides real-time speaker identification to all participants in the active session.
11. A device comprising: a microprocessor; and a computer readable medium, coupled with the microprocessor and comprising microprocessor readable and executable instructions that cause the microprocessor to execute: instructions to determine a location of a source of at least first audible information; instructions to identify an identification device at the location of the source of the first audible information, wherein the location of the identification device is determined based on different arrival times of a signal from the identification device received at a plurality of sensors, wherein the identification device is a personal communication device that comprises, within the same physical housing, a radio frequency identification tag, and wherein the identification device is identified based on a caller ID of the first party and the radio frequency identification tag in the identification device; instructions to determine an identity of the first party providing the first audible information from the identification device at the location of the source of the first audible information based on the caller ID of the first party; instructions to provide the at least first audible information as part of a conference call; and instructions to provide a first output signal identifying the first party providing the at least first audible information in the conference call.
A device identifies speakers using a processor and stored instructions. The instructions cause the processor to: determine the location of an audio source; identify a communication device (with a built-in RFID tag) at that location by matching caller ID and RFID tag, determining device location via signal arrival times at multiple sensors; determine the speaker's identity from the device's information; provide the audio as part of a conference call; and output a signal identifying the speaker.
12. The device of claim 11 , wherein the first party is associated with a first communication device, wherein a second party is also associated with the first communication device, wherein a location of the second party is determined from an identification device associated with the second party, wherein the identity of the second party is determined from the identification device associated with the second party, wherein a location of a source of second audible information is determined, wherein whether the determined location of the second audible information corresponds to the location of the second party is determined, and wherein the second party is identified as the speaking participant in response to determining that the determined location of the source of the second audible information corresponds to the determined location of the second party.
The device identifying speakers, outlined in the previous claim, can also identify multiple speakers. If a second person is also associated with the same communication device, the system can determine their location via another communication device, determine a location of a second audio source, and if the audio source and device locations match, identify the second person as the speaking participant.
13. The device of claim 12 , wherein at least a first portion of the first audible information is received while at least a first portion of the second audible information is received, wherein the identification of the first party as the source of the first audible information is provided as the first output signal and the identification of the second party of as the source of the second audible information is provided as a second output signal simultaneously.
In the multi-speaker identification device of the previous claim, the identification of the first and second speakers is performed simultaneously. The output signal identifying each speaker is generated and provided concurrently as they are speaking.
14. The device of claim 12 , wherein the first output signal identifying the first party as the party from whom the first audible signal is provided while playing a recording of the first audible signal.
In the speaker identification device described two claims ago, the output signal identifying a speaker is provided while simultaneously playing a recording of their audio. This allows for real-time association of the speaker's name with their voice.
15. The method of claim 3 , wherein the first audible signal is part of a conference call between a plurality of parties, wherein the conference call is stored as part of a transcript of the conference call, and wherein the output identifying the first party is part of the transcript of the conference call.
In the speaker identification method described in the method section 12 claims ago, the audio and speaker identification are integrated into a transcript of the conference call. The output identifying the speaker becomes part of the stored transcript for later review.
16. The method of claim 3 , wherein the first party is a visitor to an enterprise and wherein the identification device is registered when the visitor signs into an enterprise facility.
In the speaker identification method described in the method section 13 claims ago, the speaker is a visitor to a facility. The communication device used for identification is registered when the visitor signs into the facility.
17. The method of claim 3 , further comprising: registering, by a second party using voice recognition, a unique identifier associated with the second party, wherein the second party registers the unique identifier by saying the unique identifier.
The speaker identification method, described in the method section 14 claims ago, also allows a person to register a unique identifier using voice recognition. They register the identifier by speaking it.
18. The method of claim 3 , further comprising: registering, by a second party, using Radio Frequency Identification (RFID), a unique identifier associated with the second party; and in response to the second party registering the unique identifier associated with the second party, prompting for a name of the second party.
In the speaker identification method described in the method section 15 claims ago, a person registers a unique identifier using RFID. After the RFID tag is registered, the system prompts the user to enter their name.
19. The method of claim 18 , wherein the unique identifier associated with the second party is an RFID tag number and in response to the second party registering the unique identifier associated with the second party, generating a signal that states the RFID tag number to the second party along with prompting for the name of the second party.
In the RFID registration method, part of the overall speaker identification method and described in the previous claim, the unique identifier is the RFID tag number. After registering the tag, the system speaks the RFID tag number back to the person and prompts them for their name to ensure correct registration.
Unknown
October 24, 2017
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.